I

UNITED STATES DEPARTMENT OF THE INTERIOR

GEOLOGICAL SURVEY

REPORT ON TELEVIEWER LOG AND STRESS MEASUREMENTSIN HOLES USW G-3 AND UE-25pl

YUCCA MOUNTAIN, NYE COUNTY, NEVADA

By

J. M. Stock, J. H. Healy, J. Svitek, and L. Mastin

Open-File Report 86-369

Prepared in cooperation with theNevada Operations Office, U.S. Department of Energy

(Interagency Agreement DE-AI08-78ET44802)

This report is preliminary and has not been reviewedfor conformity with U.S. Geological Survey editorialstandards and stratigraphic nomenclature. Any use oftrade names is for descriptive purposes only and doesnot imply endorsement by the USGS.

Menl,o Park, California1986

Open-File Report 86-369 Open-File Report 86-369

UNITED STATESDEPARTMENT OF THE INTERIOR

GEOLOGICAL SURVEY

REPORT ON TELEVIEWER LOG AND STRESS MEASUREMENTSIN HOLES USW G-3 AND UE-25pl

YUCCA MOUNTAIN, NYE COUNTY, NEVADA

By

J. M. Stock, J. H. Healy, J. Svitek, and L. Mastin

j

CONTENTS

PageAbstract. * ............................................................ 1Acknowledgements.......................... ........................................................ 2Introduction.......................** . ... ** ***,*.. .. * *. 2Geologic Setting.......................... ......................................................... 2Borehole Televiewer - Method ........................................ 4Borehole Televiewer - Results

i. Throughgoing Fractures .......... ............... ............. 6ii. Drilling-induced Hydrofractures ............................. 11

iii. Borehole Brts........................ 13iv. Sedimentary Features........................................ 13

Hydraulic Fracturing Stress Measurements - Methodi. Horizontal Stress Values ................................... 16ii. Effect of Borehole Deviation ............................... 17

iii. Vertical Stress Values ..................................... 20Hydraulic Fracturing Stress Measurements -

Equipment and Test Chronology .......................... 21Hydraulic Fracturing Stress Measurements - Results

i. Stress Measurements in USW G-3 ............................. 26ii. Stress Measurements in Ue-25pl ............................. 26

iii. Impressions of Fractured Intervals.......................... 29Discussion

i. General Features of the Yucca Mountain Stress Field......... 30ii. Fault Stability............................................. 33

iii. Extrapolation of Results to Shallower Levels................ 35iv. Relationship to the Regional Stress Field.......e 35

Summary ............................................................. 36References .......................................................... 37Appendix I: USW G-3 Televiewer Photos............................. 48Appendix II: Ue-25pl Televiewer Photos............................. 63

ILLUSTRATIONS

Figure 1. Simplified geologic map of Yucca Mountain showing loca-tions of holes USW G-3 and Ue-25pl, as well as otherholes mentioned in text ................................ . 3

2. Sketch of borehole televiewer and determination ofstrike and dip of planar features seen in the tele-vilewer log ........ . .. . ... . . .. . .. . . . . . . 5

3. Plot of azimuth of dip (relative to geographic north)and dip, as a function of depth, of fractures seen inthe televiewer log of USW G-3............................ 7

4. Plot of azimuth of dip (relative to geographic north)and dip, as a function of depth, of fractures seen inthe televiewer log of Ue-25p............................ 8

i

Page

Figure 5. Lower hemisphere, equal area projection of poles tothroughgoing fractures seen in the televiewer log ofUSW G-3...............................................

6. Lower hemisphere, equal area projection of poles tothroughgoing fractures seen in the televiewer log ofUe-25pl .................................................

7. Tracing of USW G-3 televiewer log showing principal fea-tures visible in the upper 700 m........................

8. Example of a borehole breakout, Ue-25pl.................

9. Compilation of center azimuths of breakouts seen in theteleviewer log of Ue-25pl ...............................

10 Map view of deviation log of USW G-3....................

11. Map view of deviation log of Ue-25pl....................

12. Pressure-time records of three successful hydraulic frac-turing stress measurements in USW G-3...................

13. Pressure-time records of three successful hydraulic frac-turing stress measurements in Ue-25pl...................

14. Steady pumping pressure as a function of flow rate fortwo tests in Ue-25pl......... ........ ..........

15. Horizontal stresses plotted as a function of verticalstress, S for USW G-1, USW G-2, USW G-3 and Ue-25pl...

16. Results of stress measurements in USW G-3 and Ue-25pl,showing values of the least horizontal stress Sh, thegreatest horizontal stress SH, and the vertical stressS.*.............................. .............. .......

17. Results of stress measurements in USW G-3 and Ue-25pl,showing the critical values of Sh for which slip mighttake place on preexisting faults of optimal orientation,for values of the coefficient of friction between 0.6and 1.0 .................................................

TABLES

1. Vertical stress profile for hole Ue-25pl................

2a. Vertical stress profile for the upper part of USW G-3...

ii

9

10

12

14

15

18

19

23

25

28

31

32

34

40

42

Table

Table 2b. Vertical stress profile for the lower part of USW G-3...

3. Summary of stress measurements, USW G-3 and Ue-25pl.....

4. Lower limits on Sh from tests in the upper part ofUe-25pl .................................................

5. Constraints on hydraulic fracturing tensile strength inthe volcanic units at Yucca Mountain....................

6. Influence of borehole deviations on results of hydrofractest ...........................................

Page

43

44

45

46

47

iii

UNITED STATESDEPARTMENT OF THE INTERIOR

GEOLOGICAL SURVEY

REPORT ON TELEVIEWER LOG AND STRESS MEASUREMENTSIN HOLES USW G-3 AND UE-25P1,

YUCCA MOUNTAIN, NYE COUNTY, NEVADA

By

J. M. Stock, J. H. Healy, J. Svitek, and L. Mastin

ABSTRACT

Hydraulic fracturing stress measurements and televiewer observations weremade in drill holes USW G-3 and Ue-25pl on Yucca Mountain, Nevada, as part ofa continuing investigation of the tectonic stress field in the vicinity of aproposed site for the disposal of nuclear waste. The results from USW G-3 aresimilar to the results reported previously from USW G-1 and USW G-2 with a lowleast horizontal principal stress in a direction approximately N65 W. Thisindicates a normal faulting stress regime consistent with the inferredregional extension in this part of the Basin and Range province.

Ue-25pl was drilled on a basement high indicated by gravity and seismicrefraction data. Initial attempts to frac this hole failed because thepressures required to fracture the rock exceeded the strength of our standardpacker system. A new and stronger resettable packer system was designed,tested, and debugged during the measurement program in USW G-3 and Ue-25pl.Much of the data collected in this experiment is not used in our presentinterpretation because of obvious or suspected malfunction of this new packersystem. Based on the experience of this experiment further modifications weremade to the packer system and recent tests indicate the development of a muchimproved and stronger system. The results of this development effort will bereported separately.

The number of reliable hydraulic fracturing tests in Ue-25pl is notsufficient to reliably characterize the state of stress at this site, but somedefinite conclusions can be drawn from these data. The pressure curvesobserved during many of the frac tests are significantly different from thecurves observed In USW G-1, USW G-2, and USW G-3. The breakdown pressures arehigher, suggesting a higher least horizontal principal stress and/or a sub-stantially higher apparent tensile strength for the rocks in this hole. Theinstantaneous shut-in pressure does not clearly reveal the least horizontalprincipal stress in most of the frac tests, but two pressure versus volumepumping tests in the Paleozoic rocks show a fracture opening pressure close tothe vertical stress, suggesting that the least horizontal principal stresscould possibly be as high as the vertical stress. One good hydraulic

1

fracturing test reveals a least horizontal stress much less than the verticalstress, with a greatest horizontal principal stress close to but less than thevertical stress. No drilling-induced hydraulic fractures were observed inthis hole, in contrast to the three USW G holes which had drilling-inducedhydraulic fractures in the upper portions of the televiewer logs. Thissuggests that the Sh values are higher than the hydrostatic pressure inUe-25pl. Borehole breakouts in the lower part of Ue-25pl indicate anextension direction of N60-65W, consistent with that observed in the otherholes. These observations suggest that the stresses at Ue-25pl may stillcorrespond to a normal faulting stress regime, although it may be close to thetransition between normal and strike-slip faulting stress regimes. Ue-25plmay have more stable stresses than the other three holes, with Sh valuescloser to Sv and further from the frictional sliding criterion defined byByerlee's law.

ACKNOWLEDGEMENTS

We thank Jim Springer, Gretchen Zwart, and Keith Brown for assistancewith instrumentation and field operations, Rob Harris for analysis of densityand breakout data, and Dennis Styles, Tom Denham, Jackie Hennagan, and RickLudlum for assistance with packer operation and televiewer logging. MarkZoback helped with a valuable discussion of the results. We also thank theDOE and Fenix and Scisson personnel at the Nevada Test Site, especiallyWallace Hammer and Ray Escobedo, for their support during the field operations.

INTRODUCTION

The Nevada Nuclear Waste Storage Investigations Project has devoted sub-stantial effort to the evaluation of a prospective site for a nuclear wasterepository at Yucca Mountain, Nevada. As part of these studies, USGSpersonnel have performed borehole televiewer logs and hydraulic fracturingstress measurements in four holes at Yucca Mountain to determine themagnitudes and orientations of the horizontal principal stresses. Thisinformation is important to the evaluation of the seismic stability of therepository site and to the safe design and construction of the repositoryfacility. This report describes the results of measurements in holes USW G-3and Ue-25pl collected during February, March, November, and December 1983, andJanuary 1984. Previous measurements in holes USW G-1 and USW G-2 have beendescribed by Healy et al. (1984) and Stock et al. (1984, 1985).

GEOLOGIC SETTING

Yucca Mountain is in southern Nevada at the western edge of the NevadaTest Site (NTS) (figure 1), and is bounded by Crater Flat to the west, JackassFlats to the east, and Highway 95 to the south. Yucca Wash forms its northernboundary with the Calico Hills. Yucca Mountain comprises a thick sequence ofMiocene volcanic tuff and flow units, ranging from the Tiva Canyon member ofthe Paintbrush Tuff at the surface to the Tuff of Lithic Ridge and older tuffsat depth. These units have been tilted gently eastward in a series of faultblocks bounded by high-angle, west-dipping, normal faults that are generallyNNW- to N-striking (Scott and Bonk, 1984). The tuffs extend to at least 3 km

2

36052!3d0

US-

G-13

116°30

Figure 1. Simplified geologic mapholes USW G-3 and Ue-25pl, as wellStippled area represents alluvium;units.

Geology simplified from Scott. R.B.

1 written commun. 1984 )

of Yucca Mountain showing locations ofas other holes mentioned in the text.white outcrop is Tertiary volcanic

3

depth under the main part of Yucca Mountain (location of holes USW G-1, G-2,and G-3). The interpretation of gravity (Snyder and Carr, 1982) and seismicrefraction data (Hoffman and Mooney, 1983) ndicates a structural high in thePaleozoic subsurface beneath part of the eastern side of Yucca Mountain. Thisis also indicated at hole Ue-25pl and adjacent holes where Silurian andOrdovician carbonate units were drilled at 1,244 m depth.

USW G-3 was drilled on the crest of Yucca Mountain, completely within theMiocene volcanic units, at coordinates 360 49' 04.90" N, 116 28' 00.71" W.Its surface elevation is 1,480 m, and it has an unusually low water table, 752m below the surface. The hole is approximately vertical from the surface tothe bottom of the casing at 792 m death, then deviates to the west at agradually increasing angle,-up to 25 from vertical at the bottom of thehole. A detailed stratigraphy and description of core is given by Scott andCastellanos (1984).

Drill hole Ue-25pl, at coordinates 36 49' 38.13" N, 116 25' 20.70" W,has a surface elevation of approximately 1,120 m, and one of the shallowestwater levels seen at Yucca Mountain, 385 m. Its maximum deviation fromvertical is 8.75 at the bottom of the hole. It penetrates relativelyunfractured Miocene volcanics to 1,244 m depth, where a fault brings them intocontact with Paleozoic carbonates that continue to the bottom of the hole at1,805 m depth. Stratigraphy and core descriptions are reported by Carr et al.(1986).

BOREHOLE TELEVIEWER - METHOD

The acoustic borehole televiewer is a logging tool designed to map thesmoothness of the borehole wall (Zemanek et al., 1969, 1970). It contains a2-MHz transducer that pulses 1,800 times per second and rotates three timesper second, emitting a signal which is reflected off the borehole wall, thentransmitted uphole by wireline to be recorded at the surface. A fluxgatemagnetometer is used to orient the signal with respect to magnetic north. Thetool is pulled upward at a constant speed of 2.5 cm/s. The reflected signalis displayed on an oscilloscope with depth on the vertical axis, azimuth onthe horizontal axis, and brightness as a function of amplitude of thereflected signal. Photographs of the oscilloscope display, taken at regularintervals, are taped together to form a continuous log of the hole. Thesignal is also recorded on magnetic tape to facilitate reprocessing.

The final image is an "unwrapped" view of the inside surface of thewellbore. Strong reflected signals appear as bright areas on the televiewerlog, while zones of roughness such as fractures or breakouts disperse thereflected signals and can be seen as dark areas. A planar fracture inclinedto the borehole axis will have a sinusoidal trace on the televiewer log, fromwhich its strike and dip can be determined (figure 2). Other features visibleon the televiewer logs include borehole breakouts, large lithophysal cavities,cobbles of contrasting lithology in conglomerate beds, occasional bedding, anddrill bit- and pipe-produced wear and scuffing.

4

USWG 3 Ue- 25P1946.4 m

Kd-�

D2

1770.9 m

1778.5 m

Cn-1-

h/21

strike

… …

Th

I

NE I W

E S W N

954.0 mBOREHOLE TELEVIEWER LOG

Strike: orientation of midpointbetween peak and trough

Dip: tan 1 (h/d)

steeply dipping

fracture

shallowlY dippingbedding

andbedding- Parallel

fractures

Figure 2. Sketch of borehole televiewer and determination of strike and dip of planar featuresseen in the televiewer log. Representative portions of the televiewer logs from USW G-3 andUe-25pl are shown to illustrate steeply dipping fractures, shallowly dipping fractures, and bedding.

Because televiewer logs are plotted at a constant vertical scale, thedistortion of borehole features depends on hole diameter. USW G-3 was 10 cmin diameter, while Ue-25pl was 20 to 28 cm in diameter; thus, the horizontalexaggerations of these two logs are 3.4:1 and 6.8:1 to 10:1, respectively.This strong horizontal exaggeration makes detection of shallowly dippingfeatures very difficult, so identification of planar features (e.g., fractures,bedding) on this log is somewhat biased toward those of steeper dip.

Ideally, the televiewer should be centered i the hole during thelogging, so that the reflected signal returns directly back to theteleviewer. The televiewer log of USW G-3 was obtained in an interval whichdeviated from vertical 7 to 21 toward the west, which was too pronounced adeviation for the centralizers to keep the tool centered. As the tool wasfrequently closer to the eastern side of the hole, good signal return wasobtained from the azimuths close to east and west. However, poor signalreturn at azimuths close to north and south gave rise to the vertical blackbands present at these azimuths below 907 m (2,975'). This problem was not aspronounced in hole Ue-25pl as it was much closer to vertical.

Tool problems also may have caused the pattern of regularly spacedslanted black lines, giving the appearance of a spiral fabric in USW G-3, at890-919 m, (2,920'-3,015'), 632-946 m (2,075'-3,105'), and 968-1,013 m(3,175'-3,325'). Amplitudes of the televiewer signal and travel time crosssections show that this pattern may be caused by lateral oscillation of thebody of the tool as it is pulled upward at a constant rate and the cablepasses through exceptionally crooked parts of the hole. In repeat sections ofthe log in zones showing this pattern, the black lines occurred with the samefrequency, but in a slightly different position relative to identifiablefeatures of the borehole wall such as fractures. This pattern has not beenobserved by us in Ue-25pl, USW G-1, or USW G-2.

BOREHOLE TELEVIEWER - RESULTS

Televiewer logs were obtained from 792-1,364 m in USW G-3, and in twoseparate intervals in Ue-25pl: 480-1,202 m (Tertiary volcanic units andfanglomerate) and 1,297-1,798 m (Paleozoic carbonate units). The completelogs are reproduced in the appendices. Preexisting fractures, drilling-induced hydraulic fractures, borehole breakouts, bedding, and conglomeratecobbles are all visible in portions of these logs; their appearance andidentification are discussed below. Study of the breakouts and drilling-induced hydraulic fractures, which are stress-related, shows differences inthe response of the two boreholes to drilling and coring activities, possiblydue to differences in the stress regime between the two sites (discussedfurther below).

Throughgoing fractures

In this work, a fracture is defined as throughgoing if both updip anddowndip intersections with the borehole wall can be seen in the televiewerlog. The appendices and Figures 3 to 6 show that the distribution andorientation of throughgoing fractures differs considerably between these twoholes. The USW G-3 televiewer log does not show many throughgoing fractures;

6

AZIMUTH OF DIP DIP (DEGREES)

N EI I

S WI I

N 0 3G 60 90

800 -

900 -

1000 -

0S0

.

S

Is

9

'9

0

Bullfrogmember ofCrater FlatTuff

Tram member

of

Crater FlatTu ff

Tct

CDq:wGI-

w

0-11 00 -

bedded tuft

1200 -

1300 -

LithicRidgeTuff

TIr

THROUGHGOING FRACTURES SEEN IN TELEVIEWER LOG,USW G-3

Figure 3. Plot of azimuth of dip (relative to geographic north) and dip,as a function of meters depth, of fractures seen in the televiewer log ofUSW G-3. A simplified lithologic column is shown at the right forreference. Orientations are corrected for hole deviation.

7

THROUGHGOING FRACTURES OBSERVED IN TELEVIEWER LOG

Ue-25p# I

AZIMUTH OF DIP DIP (DEGREES)

N E S W

500

0

600-

700-

o0

do

a

o 30 s0 so. .. . 0

0

00

a00

0

0

Tcp

Teb

4,

S

Tct

8001

0.

900-

1000-

o0

0000

Trt

S0

0

P-

0

1100-

0 0

0 0 &I

Prow Pss member

of Crater Flat Tuft

Bullfrog member

of Crater Flat Tuft

Tram member

of Crater Flat Tuft

Lithic Rdge Tut

Older tuft

Fanglomerate

Older tutt

Lone MountainDolomite

Roberts MountainFormation

1I a t°

0 o1 200- o

1 300

1400

I SOO-

1600

1700!

0 7ao 't

3. ol

aCt.

.0

Coil.

jC

0

0

0

a00

sSe

e0

0

0 a

Figure 4. Plot of azimuth of dip (relative to geographic north) and dipas a function of meters depth, of fractures seen in the televiewer log ofUe-25pl. Fractures with dips less than 60 re represented by opencircles; fractures with dips greater than 60 are represented by filledcircles. Lithologic column is shown for reference. The two sections ofthe log were collected separately. The section of hole from 1,202 to1,297 m was never accessible for televiewer logging. 8

NEVADA TEST SITE USW G-3Poles to fractures seen in televiewer log

Lower hemisphere, equal area projection

N

792 - 1364 mCRATER FLAT TUFF

(BULLFROG MEMBER, TRAM MEMBER)AND LITHIC RIDGE TUFF

Figure 5. Lower hemisphere equal area projection of poles to the through-going fractures seen in the televiewer log of USW G-3. Fracture orienta-tions have been corrected for hole deviation.

9

NEVADA TEST SITE Ue-25p#1

Poles to fractures seen in televiewer log

Lower hemisphere, equal area projection

N N N

0L

488 - 873 mCRATER FLAT TUFF(PROW PASS MEMBER,BULLFROG MEMBER,TRAM MEMBER)

873 - 1202 m

LITHIC RIDGE TUFF

FANGLOMERATEAND OLDER TUFFS

1298 - 1798 mLONE MOUNTAIN DOLOMITE

ANDROBERTS MOUNTAIN FORMATION

Figure 6. Lower hemisphere equal area projection of poles to the throughgoingfractures seen in the televiewer log of Ue-25pl.

we believe that this may be due to the very thick mud coating the walls of thehole, which may have filled and smoothed all except the largest fractures.The few visible fractures are all steeply dipping and have an average strikeof N20 E, with dips to both east and west. There are more fractures visiblein the Ue-25pl televiewer log; these have fairly scattered orientations andboth steep and shallow dips. The Lone Mountain dolomite has a higher concen-tration of fractures than the other units in the hole; most of the fracturesseen in this unit have an average strike of N-S and dips from 500 to 70 ,predominantly to-the east.

Although the Ue-25pl log shows more fractures than the USW G-3 log, itshould be noted that neither of these holes have as many visible fractures asUSW G-1 and USW G-2.

Drilling-induced hydraulic fractures

'The USW G-3 televiewer log shows long fractures which are parallel to theborehole axis for many meters, along an average azimuth of N25 E (figure 7).Some of these are very Jagged, being connected to or interrupted by through-going fractures; e.g., 882-899 m (2,825'-2,950'). In other places thesefractures are fairly straight and have no throughgoing fractures associatedwith them; e.g., 975-986 m (3,200'-3,235') and 1,006-1,018 m (3,300'-3,340').Such fractures may be present elsewhere in the hole if they are within theblack bands of poor signal return resulting from the tool being off center inthe hole. These fractures, and similar ones observed in holes USW G-1 (Healyet al., 1984) and USW G-2 (Stock et al., 1984, 1985), are believed to bedrilling-induced hydrofractures. In USW G-2, these fractures are present inintervals of the hole where the corresponding core was unfractured, indicatingthat the fractures formed after that interval of the hole was drilled. Also,these fractures, where encountered, appear to follow the axis of the hole.The orientations of the straight and continuous portions indicate that thefractures are well-centered on the hole, suggesting that their orientation sstrongly controlled by the presence of the borehole and that they were formedat a later stage in the drilling process.

Drilling-induced hydrofractures are likely to form in holes where theleast horizontal stress is low relative to the vertical stress, and where thewater table is fairly far below the surface. In the presence of a lowequilibrium water table, raising the borehole fluid level to the surfacecauses a large increase in the borehole pressure above the equilibrium porepressure in the rock. If Sh is low, and this borehole pressure exceeds theleast horizontal stress on the rock, a hydraulic fracture may initiate andpropagate. Results of the hydraulic fracturing stress measurements in USWG-3, discussed below, are consistent with this mechanism for formation ofthese fractures.

Because hydraulic fractures theoretically should form perpendicular tothe least horizontal principal stress, Sh, their orientation is frequentlyused as an indicator of stress direction. In a borehole that is oblique tothe principal stress directions, these fractures would be expected to initiateat the azimuth of the least compressive stress in the borehole wall. This

11

Depth (feet)

2600 2700

I

Vi

_ A_

2800

X_

,X

_

2900

>S

X

ft

a

$

3000

I'

3100 3200

X ~A213 L

g

_\

\-

2700 2800 2900 3000 3100

Jif1-)

"I(I

I II I

3200 3300

Depth (feet)

FEATURES SEEN IN TELEVIEWER LOG, USW G-3(UPPER PORTION)

Figure 7. Tracing of USW G-3 televiewer log showing principal featuresvisible in the upper 700 m. Fractures drawn in heavier line between 3,000and 3,300 ft are inferred to be drilling-induced.

12

azimuth departs from the SH direction as SH approaches Sh and as theborehole axis approaches the Sh direction (Richardson, 1983). However, withthe observed Yucca Mountain stress field, for the deviation of USW G-3 at theposition of the drilling-induced hydraulic fractures, no more than a fewdegrees of rotation of the apparent principal stress direction would beexpected. (See the discussion of deviated holes in the following section onhydraulic fracturing theory.) Thus, the orientation of these fractures in USWG-3 is sufficiently close to SH that it can be taken to imply a N65*Worientation of Sh. No drilling-induced hydraulic fractures are seen inUe-25pl, so no directional information for that hole can be inferred by thismethod.

Borehole breakouts

Breakouts are spalled portions of the borehole wall which often appear inthe televiewer log as paired, vertical, black bands, centered approximately180 apart, with Jagged edges. When the televiewer signal is replotted withtravel time as a function of azimuth, these zones correspond to azimuths alongwhich the hole is enlarged, resulting in a roughly elliptical cross section(figure 8). They can also be identified on downhole television pictures andoriented caliper logs. It has long been known that their orientations areuseful indicators of the 5h direction (e.g., Gough and Bell, 1982; Bell andGough, 1983). Recent laboratory studies confirm that breakouts form by acombination of shear and tensile failure of the borehole wall at the azimuthof concentrated high compressive stresses, centered on the Sh direction(Mastin, 1984).

In Ue-25pl, borehole breakouts are present only in the Paleozoic units,at depths from 1,524-1,542 m (5,000'-5,060'), 1,585-1,588 m (5,200'-5,210'),1,608-1,609 m (5,275'-5,280'), 1,631-1,640 m (5,350'-5,380'), and 1,649-1,676m (5,410'-5,500'). Although the orientations of the centers of the breakoutsdo show some scatter, the average azimuth of the breakouts is N60 W (figure 9).No borehole breakouts are visible on the televiewer log of USW G-3. The blackbands at N and S azimuths in the USW G-3 log are caused by the televiewerbeing off center in the hole (as discussed above). Cross sections made byreprocessing the televiewer signal at several depths show this eccentricityclearly but provide no evidence of breakouts.

Sedimentary featuresSome sedimentary features are visible on the Ue-25pl televiewer log.

Among these are irregular light patches occurring from 1,158-1,172 m (3,800'-3,846'), with the largest at 1,168 m and 1,172 m (3,831' and 3,846') depth.These are believed to correspond to cobbles in the conglomerate unit. From1,158-1,172 m (3,800'-3,844') this unit consists of clasts of carbonate rocks,argillite, sandstone/siltstone, and tuff, in a claystone matrix (Carr et al.,1986). Some of the clasts are also visible in the downhole television log ofthe hole in this interval; their large size (exceeding 30 cm in diameter)explains their visibility on the televiewer log.

13

C"I0V-

U)

is1M N

co

ccw-w

Xa-a so

I-

E

a)Cm

T-

0U,) LO

S

NEVADA TEST SITEUE25P#1

I T WI I I

N E S W N

Cross section of borehole1527 meters

Figure 8. Example of a borehole breakout, Ue-25pl. When the televiewerlog is reprocessed to yield a cross section of the hole, the dark pairedbands in the televiewer log are found to correspond to the azimuths oflarge hole diameter.

14

N

E

elongation along

S

AZIMUTHS OF BREAKOUTS, NTS Ue-25p#1

1218 - 1524 m DEPTH

Figure 9. Compilationviewer log of Ue-25pl.

of center azimuths of breakouts seen in the tele-The average azimuth of breakouts is N60 W.

15

Bedding is generally not visible in the televiewer log of the volcanicsections of Ue-25pl and USW G-3. In the core, many of the tuff units aremassive or poorly bedded; although clear bedding is present in some parts ofthe core ("bedded tuff units"), it may be too fine or too shallowly dipping toshow up on the televiewer log. Fine laminations in parts of the Paleozoicunits in Ue-25pl (Carr et al., 1986) are also not visible in the televiewerlog; however, more massive bedding can be seen in the Lone Mountain Dolomitefrom 1,544-1,550 m (5,065'-5,085'), and in the Roberts Mountain Formation from1,771-1,779 m 5,810'-5,836'), and below 1,783 m (5,850') (figure 2). Thebeds below 1,768 m (5,800') have an averaged apparent dip of 30 toward theNNW, which, when corrected for probable hole deviation, yields true dip of38 toward N15*W. (This dip vector may be in error by up to 10 because thedeviation for this interval was estimated by extrapolation downward from thelast measurement at 1,707 m.) Bedding between 1,542 m and 1,550 m (5,060' and5,085') is less distinct, but appears to be steeper, with a strike of N25*Eand a dip of 68 NW when corrected for magnetic declination and measureddeviation of the borehole at this depth.

HYDRAULIC FRACTURING STRESS MEASUREMENTS - METHOD

Horizontal Stress Values

The hydraulic fracturing method (Hubbert and Willis, 1957; Haimson andFairhurst, 1967) directly measures the magnitude of the least horizontalprincipal stress, Sh, and indirectly yields the magnitude of the greatesthorizontal principal stress, SH. The method involves the creation of afracture in an isolated region of the borehole, and observation of itspressure-time behavior during repeated borehole pressurization cycles. A one-to two-meter interval of the borehole is sealed off between two inflatablerubber packers, and connected to the surface by a tubing string. Pressure ismeasured downhole in the tubing at the depth of the packers by a pressuretransmitter attached to a wireline, which sends signals uphole to be recordedon chart paper and on computer disk. A self-contained mechanical pressuregauge is attached below the housing of the electronic gauge to provide anadditional pressure record. A wireline lubricator at the top of the tubingprevents fluid leaks at the surface.

Fluid is pumped into the tubing from the surface to raise the pressure inthe interval between the packers until a fracture is created in the boreholewall. Once a fracture forms, pumping is stopped, the interval is "shut in".(sealed off at the surface), and the pressure decays by flow into the fractureand/or permeation into the borehole wall. The pressure at which the fracturecloses is often visible as an inflection on a plot of interval fluid pressureagainst time, and is referred to as the "instantaneous shut-in pressure" orISIP. After shut-in, the pressure normally is bled back to its pre-testvalue. Several more of these pressurization cycles are conducted to observefracture reopening and shut-in pressures. Ideally, these cycles are allconducted at the same flow rate so that comparison can be made between thebehavior of the fracture during different cycles (Hickman and Zoback, 1983).

16

After several constant-rate pumping cycles have been conducted, andfracture pumping pressure has stabilized, pumping pressures are recorded forone or two cycles of stepwise decreasing flow rates, to further constrain thepressure at which the fracture closes (e.g., Earlougher, 1977; Hickman et al.,1985). The resulting plot of pumping pressure as a function of flow rateshould be bilinear with a change n slope due to the change in apparentpermeability of the interval as the fracture closes.

This method is based on an analytic solution for stress around acylindrical hole in an elastic, isotropic medium. If the borehole is drilledparallel to a principal stress direction in homogeneous rock, theoreticalcalculations indicate that a hydrofracture should initiate at the azimuth ofleast tangential compressive stress at the borehole wall, that is, at theazimuth of the maximum horizontal principal stress, SH (Hubbert and Willis,1957). After the fracture propagates away from the zone near the borehole,the normal stress across the fracture should be equal to Sh. The boreholepressure P required to initiate this fracture is given by Haimson andFairhurst 1967):

Pb = 3Sh - SH + T - P (1)

where T is the tensile strength of the rock and P is the equilibrium porepressure in the formation. If T and P are known, the observed normalstress on the fracture can be used in equation (1) to determine SH.However, T is observed to vary greatly with sample size and testing method(e.g., Ratigan, 1983). Values of T obtained by hydraulic fracturing tensiletests in the lab must usually be extrapolated upwards to the in situ boreholesize in order to be used in equation (1). To avoid this uncertainty in T,Bredehoeft et al. (1976) developed a modified form of equation (1), using thefracture reopening pressure on later cycles, Pb (T=O):

Pb (T=O) = 3Sh - SH - Pp (2)

Equation (2) can be rewritten to solve for SH if the borehole pressure hasbeen returned to its original value by bleeding back excess pressure aftereach cycle.

Effect of Borehole Deviation

In any hydraulic fracturing test, it is assumed that the orientation ofthe fracture at the borehole is the same as that in the far-field, that is,the rock is isotropic and the (vertical) borehole is parallel to a (vertical)principal stress direction. Under these circumstances, the hydraulic fractureforms perpendicular to the Sh direction. Even for vertical boreholes, thisassumption may not always be correct because local topography and/or localstructural irregularities can influence the principal stress directions.However, in USW G-3 there is the additional complication of strong boreholedeviation (figures 10 and 11), so that the basic assumptions of the methodmust be more carefully evaluated.

17

HORIZONTAL PROJECTION OF HOLE DEVIATION, USW G-3

DISTANCE WEST (FEET)

700 600 500 400 300 200 10000

NORTH

Iv

OD

-

- z0 m

z0--I

0c-i

-.L m0 moci-

40004500

353000

4990,

Depth In feet

Figure 10. Map view of deviation log of USW G-3.

HORIZONTAL PROJECTION OF HOLE DEVIATION, Ue-25p#1

DISTANCE EAST-WEST (FEET)

60 40 20 0 20 40I

6020

NORTH

IU-

0CO

i-0z

wz

CO,0.

0

20

40

60

80

2000

3000 U

4000

3500

\- 50004500

5600Depth In feet

100 -

120-

5923

Figure 11. Map view of deviation log of Ue-25pl.

19

Regardless of the obliquity of the borehole to the stress field, thehydraulic fracture should initiate at the azimuth of least compressivetangential stress in the borehole, at a breakdown pressure equal to thisstress plus Pp - T. As the fracture propagates, it should always follow atrajectory perpendicular to the least principal stress. For an obliqueborehole, this means that the orientation of the fracture will change awayfrom the borehole. The far field normal stress on the fracture should beequal to 3 and should be the pressure corresponding to the inflection onthe pressure record when the fracture closes (ISIP).

It is possible to predict the breakdown pressures and hydraulic fractureorientations that might be expected for a borehole with the deviation observedin USW G-3 and Ue-25pl. When the observed Yucca Mountain stress field isapplied to holes of the observed deviation (Table 6), it is found that theapparent azimuths of hydraulic fractures do not differ significantly from thetrue SH azimuth. This is because the magnitudes of Sh and SH are verydifferent; as goes to zero (Sh SH), large deviations of apparent Share possible. Even with the uncertainties in the stress magnitude values usedfor the calculations in Table 6, the azimuth of the drilling-induced hydraulicfractures seen in USW G-3 should be very close to the true SH direction.For tests in more severely deviated boreholes, or in regions with horizontalstresses that are closer in magnitude, the possibility of erroneousdirectional information would be higher.

Hole deviation might influence the SH result if the actual breakdownpressure was not equal to that given in equation (1) for an ideally orientedhole. Calculations of expected breakdown pressures, when compared withbreakdown pressures for an ideally oriented borehole (Table 6), show that thesmall deviation of Ue-25pl should not have affected the breakdown pressures.However, observed Pb (T=O) pressures in USW G-3 may be 10 to 15 bars lessthan would be predicted by equations (1) and (2). Thus, use of the observedPb values to compute SH might result in a calculated SH value that was10-15 bars higher than actually present.

Vertical Stress Values

Once values of Sh and SH are obtained by hydraulic fracturing stressmeasurements, the vertical stress, Sv, is still needed to characterize thecomplete stress field. In the absence of topography, Sv is equivalent tothe weight of the overburden and can be calculated if the densities of therock units are known. Sv values for both holes (Tables 1 and 2) werecalculated using borehole compensated (BHC) density logs, and averagedensities of equivalent units from nearby holes in zones where no BHC log wasobtained. In addition, Sv in USW G-3 was corrected for borehole deviationand drift towards regions of lower surface elevation (Table 2), and correctedfor topographic effects as discussed below.

If the topography is irregular, small wavelength variations should beaveraged out prior to calculation of Sv at depth. Such corrections are mostimportant for USW G-3. For instance, the elevation at the surface of USW G-3is 1,480 m; however, the average surface elevation within a circle of 700-m

20

radius centered on the top of the hole is approximately 1,390 m. For USW G-3topographic corrections, the average elevation of the ground surface aboveeach test was obtained by integration over a diameter equal to the depth ofthe test. The contribution to Sv from the rock in the drill hole above thisaverage elevation was subtracted from the values listed in Table 2 to providea rough estimate of the magnitude of Sv at the depths of the test (Table3). Maximum uncertainty limits of these values are the uncorrected value ofSv as an upper bound, and the average topography in the absence of theexcess height of the ridge as a lower bound; this gives an uncertainty oft20-25 bars for the USW G-3 S values. An estimate for the Sv correctioncan also be derived analyticayly if surface topography can be approximated asa smooth function, and density and elastic constants are assumed uniform(e.g., Swolfs and Savage, 1985).

In hole USW H-i, a hydrology hole several km northwest of Ue-25pl (figure1), BHC density values and downhole gravimetry measurements agree very closely(Snyder and Carr, 1982), indicating that the BHC density values for this holeare representative of the region around the borehole and can be considered anaccurate source for S . Borehole gravity measurements, yielding averagedensities and interva densities, were also obtained in USW G-3 and Ue-25pl(Healey et al., 1984). Sv values based on these results are listed inTables 1 and 2. These show that below 200 m depth in Ue-25pl, the densitiescalculated by these two methods agree within 5 percent or less, and are within3 percent at the depths of our hydraulic fracturing measurements. Thedensities listed by Healey et al. (1984) for USW G-3 are more difficult tointerpret, since there is some uncertainty regarding the free air gravitygradient needed for calculation of absolute densities (Healey et al., 1984).Because borehole gravity measurements were only made in the higher part of USWG-3, they are not useful for calculation of Sv at the depths of thehydraulic fracturing tests; however, we reproduce them in Table 2 forcomparison..

HYDRAULIC FRACTURING STRESS MEASUREMENTS -EQUIPMENT AND TEST CHRONOLOGY

Measurements in USW G-3 and Ue-25pl were conducted in three separatefield operations: February-March 1983 (volcanic section of Ue-25pl), Novemberand December 1983 (USW G-3), and January 1984 (Paleozoic section of Ue-25pl).

The first set of tests, in the upper part of Ue-25pl, used a Lynessingle-set straddle packer system. In this system, the packers are inflatedthrough one-way valves connected to the fluid in the tubing string, with thetest interval between the packers sealed to excess pressure during packerinflation. These one-way valves keep the packers inflated even when a secondvalve between the tubing string and the test interval is opened to begin thehydrofrac test. During this phase of testing in e-25pl, four tests wereconducted. The packers inflated and set successfully; however, the pressurerequired to fracture the formation was in all cases much higher thananticipated. In two of these tests these high pressures caused the tubingconnection between the two packers to fail in tension before breakdown of theformation occurred. In the third test, the high pressure blew a shear plug inthe lower packer mandrel before breakdown occurred.

21

The fourth test of this series is the only one that created a successfulhydraulic fracture. During the first cycle, the tubing connecting the twopackers parted; however, both packers remained set, and a 5-cycle test,showing good breakdown and classic steady pumping pressure, was thenconducted. This test is interpreted to have induced a horizontal fracture,because the recorded shut-in pressure was almost identical to the expectedvertical stress. This does not mean that the vertical stress is the leastprincipal stress, however; creation of horizontal fractures is favored if thepackers are not held together by the tubing connection (which normally acts toreduce the vertical shear across the interval). This test is, therefore, notdiscussed further in this report. The three failed tests provided informationon lower limits for Sh (discussed below).

The many test failures in the volcanics in Ue-25pl were primarily due tothe unexpectedly high test pressures needed and the relatively large boreholesize which reduced the strength of the packers. Thus, after consultation withpersonnel in the drilling program at the Nevada Test Site, we decided todevelop a new packer system for future work, and a contract was negotiatedwith TAM International to design and build a new packer system. The newhigh-pressure resettable packer design has two advantages over the Lynespackers. First, the circulation valve that opened the tubing to the testinterval and the valves to the packer elements could be reset individuallydownhole, making it unnecessary to pull the packers after each test and thussaving several hours per test. Second, the connections between the packers,and the packer elements, were built to withstand much higher pressures, makingit possible to test in higher stress regimes. A third mechanical valve ("dumpvalve") at the top of the packer assembly was used to equalize the pressuresbetween the tubing and the annulus prior to deflating the packers. Theresettable valves were operated by vertical movement and rotation of the drillsteel. In a vertical hole, the operator could feel the operation of thevalves while turning the pipe with a pipe wrench, but in more deviated holesthis required practice and considerable skill. Initially, the settingmechanism of these packers proved difficult to lock into the inflate positionin impermeable rock, due to a hydraulic lock condition. This design problemwith the prototype model caused most of the tests in USW G-3 and the Paleozoicunits of Ue-25pl to fail. The packer has since been modified by TAM to avoidthis problem.

During the second field operation, 13 tests were attempted in USW G-3 atdepths between 1,060 m and 1,356 m. Of these, only three were successful(1,074 m, 1,338 m, and 1,356 m; figure 12). Eight tests failed because thepackers would not remain inflated. In the remaining two tests, the packersset properly the first time, and good breakdowns were observed, indicatingthat a hydraulic fracture was initiated. However, attempts to use the dumpvalve to return the pressure in the frac interval to its pre-test value beforethe second pumping cycle caused the packers to deflate, so the tests could notbe completed.

In USW G-3, which had a water level of 752 m subsurface, bleeding offexcess pressure at the urface would only return the borehole pressure tohydrostatic pressure, 74 bars above the equilibrium pore pressure. Theborehole pressure could be returned to its equilibrium value at the end of

22

400-z U SW G-3

200-L 0 o - | | 1074 m

150- expectedIW surfaceI 100 , pressure -S H

0O <50 81i

0 0 20 40 60 80 100 120 140

TIME (MIN)

Z 400

o2 200

150 USW G-3 1356 mpacker resety surface

packer -pressure

unset S~

M 100C

w 2Pe 50 eco

0.

0 20 40 60 80 10 120 10 160 10 20 220TIME (MIN)

400Z3-200- USW G-3 1338 m

0%._Ijold. 0

se 160 surface - SHpressure - S h

6 50 pPa.

00 20 40 60 80 100 120 140 160 180 200 220 240

TIME (MIN)

Figure 12. Pressure-time records of three successful hydraulic fracturingstress measurements in USW G-3. Lower curve in each frame is a copy ofthe record of the downhole mechanical pressure gauge. Arrows indicate thetime when the circulation valve was opened and the hydraulic fracturingtest began. The upper field for each test indicates the approximate rateof fluid flow at the surface, calculated by counting stroke rate of thedrill rig mud pump used for testing.

23

each cycle, either by opening the dump valve or by swabbing the pipe. Swabbingwas attempted several times without success. In two of the successful testsin USW G-3, the dump valve was not used, so fluid pressure was bled back tosurface hydrostatic pressure and then allowed to decay slowly by permeationbetween cycles. In the third test, the dump valve worked properly, and thepressure returned to pore pressure in the hole between cycles.

Previous testing in USW G-1 and USW G-2 indicated that pressures close tosurface hydrostatic pressure (i.e., the hydrostatic pressure of a column ofwater from the surface to the depth of interest) would often cause fracturingof the borehole wall due to the low magnitudes of the horizontal stresses(Healy et al., 1984; Stock et al., 1984, 1985). The presence of drilling-induced hydraulic fractures in USW G-3 suggested that the horizontal stresseswere likely to be low there as well, so packer inflation pressures for thetests in USW G-3 were set to be lower than surface hydrostatic pressure.

In the third field operation, ten tests were attempted in the Paleozoicunits of Ue-25pl. Because of the previous problems with manipulation of thedump valve, and difficulties with the pressure seals, the dump valve wasomitted from the packer assembly for these tests. The same difficulty withthe packer setting mechanism, however, caused five of these tests to fail. Intwo other tests the packers set properly, but the test intervals leaked. Inthe three remaining tests, discussed below, the packers set properly andborehole pressure was bled back to surface pressure between cycles (figure13). No swabbing was attempted. We used packer inflation pressures higherthan surface pressure in all the tests in Ue-25pl because our earlier tests inthe upper part of this hole suggested that the horizontal stresses andbreakdown pressures were generally higher than observed in USW G-1 and USW G-2.

The tests in the volcanics in Ue-25pl were done with a Racine C-1l pumpwith a pumping capacity of 0 to 50 L/min. This pump has pressure and volumecharacteristics best suited to most of our hydraulic fracturing operations.For the tests in USW G-3, the drill rig mud pump was used for all the testsbecause flow rates higher than 50 L/min were desired. Flow rates werecalculated by counting piston strokes; the uncertainty in these flow rates isestimated to be within about 10 . Total volume pumped was recorded on adigital flow meter and reset after each cycle.

Tests in the Paleozoic section of Ue-25pl were begun using the RacineC-1l hydrofrac truck pump. This pump failed, and two air-driven pumps thatwere tried next also failed, so a Halliburton service truck was brought in tofinish the testing. High flow rates were used for the remainder of thesetests, and the actual flow rate at any time must be estimated from the pumpspeed under no load. The flow rates indicated for the fracs in Figures 12 and13 are based on counting piston strokes and may have varied slightly from thevalues indicated, especially as pumping pressure increased.

24

f Z-Z_ 4 0 0

-Zts 2000I 0

U. 0

c

w4

M

%O0

Cwx.

500

400

300

200

100

0

Z 4C-z

3 2C

LO

0D

wM

%O

:

CD

cw0..

26

20

151

10

6

0 20 40 60 80 100 120 140 160 180 200 220 240

TIME (MIN)

o -

0-

0packer et Ue-25p*+1 1573 m0

I0 r - @ l> -surface pressure

0 40 80 120 100 200

TIME (MIN)

I

O _z'.Z

0..I

C0

_Ccow

U

:

Il

a-

I

TIME (MIN)

Figure 13. Pressure-time records of three successful hydraulic fracturingstress measurements in Ue-25pl. Symbols and sources of data are the sameas in Figure 11.

25

HYDRAULIC FRACTURING STRESS MEASUREMENTS - RESULTS

Stress Measurements in USW G-3

The three tests n USW G-3 (figures 11 and 13, and Table 3) showed veryconsistent values of the least horizontal principal stress Sh. For the test,at 1,074 m, in the Tram member of the Crater Flat Tuff, Sh was determinedfrom the behavior of the fracture during all three cycles. Cycle 1 showed agood (although surprisingly low) breakdown pressure, and cycles 2 and 3 showedrepeatable fracture reopening and pumping pressures. The similarity of thepressure records from these two cycles shows that the fracture stabilizedsufficiently, so this is a good value of Sh.

During the test at 1,356 m, in the Lithic Ridge Tuff, the packersdeflated so only two pressurization cycles were obtained. Cycle 1 shows asharp breakdown pressure, and cycle 2 a clear fracture reopening pressure.The value of Sh is taken from cycle 2, but if more cycles had been conducted,this value might have decreased slightly as fracture behavior stabilized.Thus, although 115 bars is probably close to the Sh value for this depth, itmay be a few bars too high.

In the test at 1,338 m in the Lithic Ridge Tuff, the first cycle shows agood breakdown, and the second cycle has a clear fracture reopening pressure.However, halfway through cycle 2 the pressure dropped dramatically, and thefracture pressurization during reopening on cycles 3 and 4 is much slower thanon cycle 2, even though the same high pumping rates were used, possibly indi-cating that the fracture broke around the packers during cycle 2, so thatcycles 3 and 4 are pumping tests communicating with the open hole. The Shvalue of 114 bars from this test is thus based on cycles 1 and 2. As discussedabove, the fracture behavior may not have completely stabilized by this time,so this value of Sh may be slightly high.

Values of the greatest horizontal stress, SH, were computed usingequation (2) and the fracture reopening pressures on the second cycle (Table3). These values are believed to be reasonable in spite of the boreholedeviation (see discussion of equation 1). For the tests at 1,074 m and 1,338m, even though the excess pressure was not completely bled back between cycles,the short time elapsed between cycles and the very thick mud cake on the bore-hole wall probably prevented significant increase in P of the surroundingrock between cycles 1 and 2. Thus, equilibrium values of P are used in thecomputations. For all three tests, the SH values are considerably less thanSv, corresponding to a normal faulting stress regime with = 0.5, where -

(92- S3)/(S1 - S3) (Angelier, 1979), with S1 S and S3 = Sh for a normalfaulting stress regime.

Stress Measurements in Ue-25pl

The test at 1,564 m in the Lone Mountain Dolomite consisted of two cyclesrun at a flow rate of 150 L/min, one cycle at a flow rate of 400 L/min, andone cycle with rates from 40 to 400 L/min to measure the normal stress on thefracture (figure 13). Pressurization rates were similar on the first three

26

cycles, and no breakdown pressure was seen, suggesting that a preexistingfracture was probably being tested. Because the test pressures approachedvalues equal to twice the packer inflation pressure, the packers were unsetand inflated again at a higher pressure. Two more cycles were conductedshowing very similar behavior to the previous cycles. Plots of pumpingpressure as a function of flow rate indicate that the normal stress on thefracture is 337 i 11 bars (figure 14). Because this test may have opened apreexisting fracture of unknown orientation, the normal stress on the fractureis only used as an upper bound on Sh (figure 13). This stress is close tothe vertical stress Sv. Although no fractures were identified on the tele-viewer log of this interval prior to hydrofrac testing, bedding parallelfractures, which are nearly horizontal, are present elsewhere in the tele-viewer log of the Lone Mountain Dolomite.

The second test in the Lone Mountain Dolomite, at 1,573 m depth, had abreakdown of 236 bars on the first cycle, and a fracture reopening pressure ofabout 195 5 bars on the second and third cycles. Several different pumpingrates were used on the third cycle in an attempt to obtain steady pumpingpressures and further constrain Sh. The lower pressure observed at very highflow rates at the end of the third cycle suggest that the fracture broke aroundthe packers and was then connected to the open hole. The Sh value of 207bars from this test is thus based only on observation of the shut-in curvesfrom cycles 1 and 2 and the beginning of cycle 3.

The third test,-at 1,693 m in the Roberts Mountain Formation, was similarto the test at 1,564 m in that no distinct breakdown was observed. The firstand second cycles look very similar except that all pressures are slightlyhigher during cycle 2. This may be due to a slight increase in flow rateduring cycle 2. The value of 365 bars given for the normal stress on thefracture is based on the steady pumping pressures and flow rates recordedduring cycle 3 (figure 14). Because this test may have reopened a preexistingfracture, this value is used as an upper bound for S3. (We assume, based onthe other tests in Ue-25pl and USW G-3, that Sh = S3.) The normal stresson the tested fracture is close to Sv. Again, no fractures were observed inthe televiewer log of this interval, taken prior to testing.

Because of the unusual character of the tests at 1,564 m and 1,693 m, weare not confident in estimating H. SH for the test at 1,573 m was computedusing equation (2) and the fracture reopening pressure on cycle 2 (figure 13).This value of SH may be a few bars too high because P may have increasedslightly between cycles 1 and 2, due to the previously-noted effect of excesspressure remaining in the borehole between cycles. No mud coated the boreholewall in Ue-25pl to retard permeation of excess pressure into the formation;however, since only a short time elapsed between cycles 1 and 3, there would beonly a slight effect on the magnitude of P We prefer to use equation (2)with its associated uncertainty rather thag equation (1), because we have notensile strength information on the Lone Mountain Dolomite. This results in avalue of 310 * 11 bars for the test at 1,573 m depth, lower than the predictedSv value of 356 bars. These stresses are close to the transition betweennormal and strike-slip stress regimes, with Sh < SH < Sv.

27

¢O 400- . 0

Mz 300 6

.,

a. 30

£ 2000

100 ~NTS Ue-25p#1

E. 100 L16

1564 m

1 00 200 300 400

_ 3~~~~~~~~~~~

CO~~~~~~~F 0 2 T LMN

Cr 3 0F<gure 1

w 300-

a. 200

NTS Ue-25p#te

100 ~~1693 m

0 ~100 200 300 400

FLOW RATE (L/MIN)

Figure 14. Steady pumping pressure as a function of flaw rate for twotests in Ue-25pl. Numbers next to the points indicate the pumping cycleduring which the data were obtained. The change in slope of the lineindicates the change in apparent permeability of the interval as thefracture closes. The normal stress on the fracture is determined byextrapolation of the open-fracture gradient back to zero flow rate.

28

The three tests in the volcanics in the upper part of Ue-25pl, in whichthe packer system failed prior to breakdown of the formation, can be used toplace lower bounds on the magnitude of Sh (Table 4). From equation (1) andthe inequalities Pfailure < Pb and Sh < SH, the following relation canbe derived:

Sh < Pfalure T + P (3)2

Since no laboratory values of T are available, an average value for T of 54*10bars has been used in equation (3). This number is based on all availablevalues of T derived from hydraulic fracturing tests in the volcanics at YuccaMountain (Table 5), using the equation

Thydrofrac = Pb - Pb (T0)

after Bredehoeft et al. (1976).

The calculated lower bounds on Sh (Table 4) unfortunately do notprovide significant new constraints on the Sh magnitude at these depths.These lower bounds are very low, falling close to the frictional stabilitylines of g = 0.6 and = 1.0. Sh values at these depths, if close to theselower bounds, would be close to the frictional stability line, consistent withthe good Sh value obtained at 1,573 m in the hole. In this case, the stressstate in the upper part of Ue-25pl may not have been very different from thatencountered in the equivalent units in the three USW G holes. The higherbreakdown pressures in Ue-25pl could have resulted from a slightly highervalue of Sh and slightly lower value of S, which would considerablyincrease the required breakdown pressure and at the same time make the stressstate slightly more stable. Alternatively, Sh values could be much higherthan the lower bounds; the uncertainties do not allow this to be resolved.

Impressions of Fractured Intervals

Normally, at the end of hydraulic fracturing stress measurements, orientedimpression packers are used to determine the'orientation of the hydraulicfractures and hence the Sh'direction. Due to time constraints, impressionswere taken in only two intervals in USW G-3: at 1,059 and 1,338 m. Both ofthese impressions showed high angle fractures with apparent dips (in theborehole reference frame) of 86' and 84', respectively. The observedorientations of these fractures, when corrected for deviation of the borehole,yield fracture dips of 73 toward S74 W (at' 1,059 ml and 69' toward N60OW (at1,338 m). This implies Sh directions of S750W * 10 and 60'W 10",respectively. The latter direction is reasonably consistent with theorientations derived from breakouts and drilling-induced hydrofracturesdiscussed above. The S75*W direction is anomalous, especially in view of thefact that it is the shallower of the two. The 20 dip of both of thesefractures toward the west is opposite to the direction of rotation that mightbe predicted from topographic rotation of the principal stress directions.There is some loss of accuracy in the instrument orientations due to theborehole deviations of 140 and 190 in these intervals, which exceed the 10-range of the Kuster single shot PeeWee magnetic survey instrument used. The

' 29

excess borehole deviation requires interpolation of the instrument's readingsbeyond scale, and some interference between the compass and its housing mayhave occurred.

Only one impression was taken at site Ue-25pl due to time limitations.The impression was damaged by material which fell in on the packer andscrubbed over the impression material as the tool was brought out of the hole,so that no usable impression was obtained.

DISCUSSION

General Features of the Yucca Mountain Stress Field

The results from USW G-3 agree with previous measurements from USW G-1and USW G-2, in which the magnitudes of both the least horizontal stress Shand the greatest horizontal stress SH are less than the vertical stressSv, corresponding to a normal faulting stress regime (Sh < SH < S)(figure 15). One good test from Paleozoic rocks in Ue-25pl is also inagreement with this general pattern. Two other tests above and below thisgood test constrain Sh to be less than S . The Sh direction of N65*W i10 (obtained frgm the orientation of drilling-induced hydraulic fractures nUSW G-3) and 60 W (from the borehole breakouts in Ue-25pl) agrees veryclosely with the Sh orientation of N65'W * 10 obtained from similarfeatures in USW G-1 (Healy et al., 1984) and USW G-2 (Stock et al., 1984).Thus, WNW-directed extensional stress is consistently observed in the drillholes we have tested at Yucca Mountain.

Within this general pattern, there are apparent, individual differencesamong the drill holes. In the three USW G holes, SH was approximatelyhalfway between Sh and S in magnitude ( 0.5). Sh was very lowrelative to Sv so that f the borehole pressure was raised by filling thehole to the surface with fluid, the borehole pressure would exceed Sh andhydraulic fractures would be likely to extend and propagate (figure 16). Inthe Paleozoic section of Ue-25pl, however, SH was almost as high as Sv (I= 1.O), so that the stress regime was close to the transition between normaland strike slip (Sh < SH < Sv). Because Sh is higher relative to Svin Ue-25pl than it is in The three USW G holes, filling the borehole to thesurface with fluid would not cause the borehole pressure to exceed Sh*Therefore it is not surprising that drilling-induced hydrofractures are notseen in the Paleozoic section of Ue-25pl.

Lower bounds on the least horizontal principal stress can be calculatedfrom tests in the volcanics in Ue-25pl where the packers failed beforebreakdown. These lower bounds show that the stress regime there may have beenone of normal faulting, similar to the stress state in the three USW G holes,with slightly more stable stresses. The lack of drilling-induced hydraulicfractures and breakouts in the volcanic units of Ue-25pl also suggest that thehorizontal stresses are closer in magnitude there, and with Sh closer toSv It is possible, although unlikely, that these differences between thevolcanic units in Ue-25pl and in the USW G holes may be enhanced by thedifferent drilling techniques used. The three USW G holes were drilled with

30

HORIZONTAL STRESS (BARS)

0 100 200 300 400

0 * I I

USW 0-1o USW G-2o USW 0-3

0 Ue25ps1

dc 100 Open symbols S

100 Filled symbols Sh

Arrows Indicate upper bounds

0 .24

u1

300 \

400

Figure 15. Horizontal stresses plotted as a function of vertical stress,S., for USW G-1,.USW G-2, USW G-3, and Ue-25pl. Open symbols are SHvalues, and filled symbols are Sh values. Dashed lines show thecritical values of Sh for which slip might take place on preexistingfaults of optimal orientation for values of the coefficient of frictionbetween 0.6 and 1.0. Bars on USW G-3 values are uncertainty limits onSv values.

31

STRESS (BARS)

Sh

e',co

w 1000

Iwz

w 15000

2000

a

Iq: 500

a-wU 1000a

1500

TERTIARYVOLCANICS

CONGLOMERATESAND

VOLCANICS

LONE MTNDOLOMITE

ROBERTSMTN FM

PAINTBRUSH TUFF

CRATER FLATTUFF

TRAM MEMBER

LITHIC RIDGE

BEDDED TUFF

USW G-3

Sh

SH

N

Figure 16. Results of stress measurements in USW G-3 and Ue-25pl showingvalues of the least horizontal stress Sh, the greatest horizontal stressSH, and the vertical stress S Bars indicate the uncertainty in theSH values. Two tests in Ue-29pl have arrows to indicate that the valuesare upper limits on Sh (see discussion in text). Dashed line indicatesthe pore pressure that would result if the boreholes were filled to thesurface with fluid during drilling.

32

mud and continuously cored, whereas the volcanics in Ue-25pl were drilledmainly with air soap and not cored continuously. However, the much higherbreakdown pressures required to fracture the volcanics in e-25pl are a clearindication that the stress state must be slightly different there.

Fault Stability

In a given normal faulting stress field, the preexisting fault mostlikely to slip ("favorably oriented preexisting fault") is one striking alongthe SH direction, and having a dip e, where 2e = (tan-l + 90*), and isthe coefficient of friction of the fault plane. For such a fault, the slipcondition can be written in the form

(Sv - Pp)/(Sh - Pp) = (p2 + 1)1/2 + ]2 (4)

(e.g., Zoback et al., 1978; McGarr-et al., 1982). If the ratio of(Sv-Pp)/(Sh-Pp) is less than this function of , the fault will bestable and no frictional sliding should occur; however, if the stress ratioexceeds this value, the fault can be expected to slip.

Because favorably oriented preexisting faults (strike N250E, dip 60-670)exist at Yucca Mountain (figure 1 and Scott and Bonk, 1984), we use equation(4) to obtain critical values of Sh as a function of Sv and . The linesplotted in Figure 17 show that for USW G-3, the measured Sh values are closeto those at which slip might be expected to occur on favorably orientedpreexisting faults for reasonable values of . The Sh value measured inUe-25pl is not quite as close to the critical value at which slip would beexpected to occur. This is consistent with evidence from the occurrence ofbreakouts and drilling-induced hydraulic fractures that it may be experiencingsomewhat higher horizontal stresses which result in a more stable stressregime. Bounds on Sh from tests in the volcanics in the upper part ofUe-25pl also suggest that favorably oriented preexisting faults may be lesslikely to slip at this part of Yucca Mountain. However, since no successfulmeasurements were obtained in this part of Ue-25pl, it would be advisable tomake measurements in other holes nearby rather than rely on extrapolation ofthese results from the lower portion of the hole.

An alternative model that frequently has been used to predict stressdistribution with depth is one of gravitational loading. If a laterallyconfined, homogeneous rock mass is subject to a vertical gravitational force,then S will be equal to the weight of the overlying rock. Plane strainanalysis predicts that the horizontal stresses will be equal, with a magnitudeof v/(l-v)]S , where v is Poisson's ratio (e.g., Jaeger and Cook, 1979).Such a model Ys in reasonable agreement with the Sh values obtained from theUSW G holes (Swolfs and Savage, 1985). However, this model does not success-fully explain the Sh result from Ue-25pl, nor the high values of SH withrespect to Sh that are observed throughout the area, from both our stressmeasurements and other regional stress indicators. We feel that there aredifficulties in applying this model to the NTS area for the following reasons.

33

STRESS (BARS)

500

LUwM

I--

Ia.w0

Ue-25p 1*-+ lower bound on Sh

<-o upper boundon Sh

p( *Sh

< Xtw 19S H1

1

2000

500

USW G-3

I--U1L

a-W0

500

1000

* Sh

S H

1500, \ 'SW

&, O

'

Figure 17. Stress values plotted against depth in USW G-3 and Ue-25plshowing values of the least horizontal principal stress Sh and thegreatest horizontal principal stress SH. Dashed lines indicate thecritical values of Sh for which slip might take place on preexistingfaults of optimal orientation for values of the coefficient of frictionbetween 0.6 and 1.0. Lithostatic curve S in USW G-3 is not correctedfor topographic effects.

34

(1) Such a model predicts that the two principal horizontal stresses willbe equal in magnitude (.e., =O). This implies considerable scatter in theSh and SH directions. Such scatter has been observed in other tectonicareas, but not in the TS area. Swolfs and Savage (1985) suggest that strongfracture anisotropy may control topography and contribute to higher SHvalues. However, consistent Sh directions are obtained from regional stressdata from a variety of depths in the region, including areas lacking thetopographic grain and strong fracture anisotropy present at shallow depths atYucca Mountain.

(2) The assumption of lateral confinement as a far field boundarycondition may not be realistic for the Yucca Mountain area, as the region isseismically active and undergoing measurable geodetic strain.

(3) A gravitational-loading model in a region with a surface water tableis consistent with fault stability at all depths. However, in a region ofvery low water table, it predicts slip on preexisting faults at depth, wherethe [v/(1-v)]S Shline crosses the yerlee's law stability line givenby equation (4yf. The instability of these faults would increase with depth.Microseismicity suggests that Yucca Mountain is relatively stable, so stressesdue to pure gravitational loading at depth seem unlikely.

Extrapolation of Results to Shallower Levels

All of the successful hydraulic fracturing stress measurements obtainedat Yucca Mountain were at depths below that of the repository block. Theseresults are of use in evaluating the regional stress field, but may not bedirectly extrapolated to shallower levels, because of the increased importanceof local topographic variation near the surface. It seems vital to performshallower stress measurements (both hydraulic fracturing and overcoring), aswell as examine all available borehole data that may be pertinent to the stressfield. In particular, televiewer logs, downhole television, geophysical logs,and cores may reveal useful information on fracture density, fracture orienta-tion, and the presence and orientation of drilling-induced hydraulic fracturesand breakouts at shallower levels of Yucca Mountain.

Swolfs and Savage (1985) present an analytic solution for the near-surface stress distribution at Yucca Mountain by approximating the topographyas a symmetric ridge. The ridge shape results in high lateral stresses andconsequent increased fault stability within the first few hundred meters ofthe surface, which do not strongly depend on the boundary conditions of themodel at depth. Our stress measurements are too deep to provide a check onthese predicted high near-surface stresses. This model merits furtherinvestigation by shallower stress measurements, as it would have importantimplications on the stress field at repository depths.

Relationship to the Regional Stress Field

The stress regime at Yucca Mountain fits the pattern observed elsewherein regions of active faulting, where the magnitudes of the principal stressesappear to be controlled by the frictional strength of preexisting faults, as

35

described by equation (4) above (Brace and Kohlstedt, 1980; Zoback and Healy,1984). Although recent active faulting has not been documented at YuccaMountain itself, it is present in the surrounding area. Recent earthquakefocal mechanisms from the TS area are generally strike-slip with minorcomponents of normal motion (Rogers et al., 1983). The observed focalmechanisms would not be inconsistent with slip on N- to E-striking preexistingfaults, due to a stress regime similar to that observed at Yucca Mountain(Sh < SH < Sv, with Sh oriented N65 W and = 0.5 to 0.7). However,these focal mechanisms can also be fit by a stress field with similarhorizontal stress orientations but with SH > S. Thus, it is possiblethat the SH values from Yucca Mountain are slightly lower than thoseexpected elsewhere in southern Nevada.

SUMMARY

Hydraulic fracturing stress measurements in drill holes USW G-3 andUe-25pl at Yucca Mountain demonstrate that the stress regime is generally oneof normal faulting, with Sh < SH < Sv. The least principal stress(Sh) direction is N60*W - N70 W, as determined from the orientation ofdrilling-induced hydraulic fractures visible in the USW G-3 televiewer log,and from borehole breakouts visible in the Ue-25pl televiewer log. In USWG-3, Sh values are much lower than Sv values, and are close to values forwhich frictional sliding might occur on favorably oriented preexisting faults,for values of the coefficient of friction between 0.6 and 1.0. SH in USWG-3 is roughly intermediate in magnitude between Sh and Sv. In Ue-25pl,one good test indicates that Sh is less than S, and SH s closer inmagnitude to Sv, so that the stresses are near the transition between normaland strike-slip stress regimes. Two other tests constrain Sh to be lessthan or equal to Sv This difference in stress regimes between the twoholes may explain the observed variation in occurrence of stress-relatedfeatures such as borehole breakouts (seen only in the lower part of Ue-25pl)and drilling-induced hydraulic fractures (seen in much of USW G-3).

Stress results from USW G-3 are very similar to results from previouswork in holes USW G-1 and USW G-2 at Yucca Mountain. Considered together, thestress measurements from the four Yucca Mountain holes present a consistentpicture of WNW-directed extension, with stresses that generally form a normalfaulting stress regime but in some areas, such as near Ue-25pl, may approachthe transition between normal and strike-slip regimes. These stresses are notinconsistent with observed focal mechanisms from the Nevada Test Site area,suggesting that they could be regionally representative.

36

REFERENCES

Angelier, J., 1979, Determination of the mean principal directions of stressesfor a given fault population: Tectonophysics, v. 56, p. T17-T26.

Bell, J. S. and D. I. Gough, 1983, The use of borehole breakouts in the studyof crustal stress, in Hydraulic Fracturing Stress Measurements, NationalAcademy Press, Washington, D.C., M. D. Zoback and B. C. Haimson, eds., p.201-209.

Brace, W. F. and D. L. Kohlstedt, 1980, Limits on lithospheric stress imposedby laboratory experiments, Journal of Geophysical Research, v. 85, p.6248-6252.

Bredehoeft, J. D., R. G. Wolff, W. S. Keys, and E. Shuter, 1976, Hydraulicfracturing to determine the regional in-situ stress field, PiceanceBasin, Colorado: Geological Society of America Bulletin, v. 87, p.250-258.

Carr, M. D. et al., 1986, Geology of drill hole UE-25pl: A pre-Tertiary testhole at Yucca Mountain, southern Nye County, Nevada: U.S. GeologicalSurvey Open-File Report, in preparation.

Earlougher, R. C., Jr., 1977, Advances in well test analysis: AIMEMonograph 5, American Institute of Mining, Metallurgical, and PetroleumEngineers, Dallas, Texas.

Gough, D. I. and J. S. Bell, 1982, Stress orientation from borehole wallfractures with examples from Colorado, East Texas, and northern Canada:Canadian Journal of Earth Sciences, v. 19, p. 1358-1370.

Haimson, B. C. and C. Fairhurst, 1967, Initiation and extension of hydraulicfractures in rock: Society of Petroleum Engineers Journal, v. 7, p.310-318.

Healey, D. L., F. G. Clutsom, and D. A. Glover, 1984, Borehole gravity metersurveys in drill holes USW G-3, Ue-25pl and Ue-25cl, Yucca-Mountain area,Nevada: U.S. Geological Survey Open-File Report 84-672.

Healy, J. H., S. H. Hickman, M. D. Zoback, and W. L. Ellis, 1984, Report onteleviewer log and stress measurements in core hole USW-G1, Nevada TestSite: U.S. Geological Survey Open-File Report 84-15.

Hickman, S. H., J. H. Healy,'and M. D. Zoback, 1985, In situ stress, naturalfracture distribution, and borehole elongation in the Auburn geothermalwell, Auburn, New York: Journal of Geophysical Research, v. 90, p.5497-5512.

Hickman, S. H. and M. D. Zoback, 1983, The interpretation of hydraulicfracturing pressure-time data for in-situ stress determination, inHydraulic Fracturing Stress Measurements, National Academy Press,Washington, D.C., M. D. Zoback and B. C. Haimson, eds., p. 44-54.

37

Hoffman, L. R. and W. D. Mooney, 1983, A seismic study of Yucca Mountain andvicinity, southern Nevada: Data report and preliminary results: U. S.Geological Survey Open-File Report 83-588.

Hubbert, M. K. and . G. Willis, 1957, Mechanics of hydraulic fracturing:Journal of Petroleum Technology, v. 9, p. 153-168.

Jaeger, J. C. and N. G. W. Cook, 1979, Fundamentals of Rock Mechanics: NewYork, Chapman Hall.

Mastin, L. G., 1984, The development of borehole breakouts in sandstone:Masters Thesis, Stanford University, Stanford, CA.

McGarr, A., M. D. Zoback, and T. C. Hanks, 1982, Implications of an elasticanalysis of in situ stress measurements near the San Andreas fault:Journal of Geophysical Research, v. 87, p. 7797-7806.

Ratigan, J. L., 1983, A statistical fracture mechanics determination of theapparent tensile strength in hydraulic fracture, in Hydraulic FracturingStress Measurements, National Academy Press, Washington, D.C., M. D.Zoback and B. C. Haimson, eds., p. 159-166.

Richardson, R. M., 1983, Hydraulic fracture in arbitrarily oriented boreholes:an analytic approach, in Hydraulic Fracturing Stress Measurements,National Academy Press, Washington, D.C., M. D. Zoback and B. C. Haimson,eds., p. 167-175.

Rogers, A. M., S. C. Harmsen, W. J. Carr, and W. Spence, 1983, Southern GreatBasin seismological data report for 1981 and preliminary data analysis:U. S. Geological Survey Open-File Report 83-669.

Scott, R. B. and J. Bonk, 1984, Preliminary geologic map of Yucca Mountainwith geologic sections, Nye County, Nevada: U.S. Geological SurveyOpen-File Report 84-494.

Scott, R. B. and M. Castellanos, 1984, Stratigraphic and structural relationsof volcanic rocks in drill holes USW GU-3 and USW G-3, Yucca Mountain,Nye County, Nevada: U. S. Geological Survey Open-File Report 84-491.

Snyder, D. B. and W. J. Carr, 1982, Preliminary results of gravityinvestigations at Yucca Mountain and vicinity, southern Nye County,Nevada: U.S. Geological Survey Open-File Report 82-701.

Stock, J. M., J. H. Healy, and S. H. Hickman, 1984, Report on televiewer logand stress measurements in core hole USW G-2, Nevada Test Site, October-November, 1982: U.S. Geological Survey Open-File Report 84-172.

Stock, J. M., J. H. Healy, S. H. Hickman, and M. D. Zoback, 1985, Hydraulicfracturing stress measurements at Yucca Mountain, Nevada, and relation-ship to the regional stress field: Journal of Geophysical Research, v.90, p. 8691-8708.

38

Swolfs, H. S. and W. Z. Savage,at Yucca Mountain, Nevada:Symposium, Rapid City, SD,

1985, Topography, stresses, and stabilitypaper presented at the U. S. Rock Mechanics

June 1985.

Zemanek, J., R. L. Caldwell, E. E. Glenn, S. V. Holcomb, L. J. Norton, and A.J. D. Straus, 1969, The borehole televiewer - a new logging concept forfracture location and other types of borehole inspection: Journal ofPetroleum Technology, v. 21, p. 702-774.

Zemanek, J., E. E.evaluation by35, no. 2, p.

Glenn, L. J. Norton, and R. L. Caldwell, 1970, Formationinspection with the borehole televiewer: Geophysics, v.254-269.

Zoback, M. D. and J. H. Healy, 1984, Friction, faulting, and in-situ stress:Annales Geophysicae, v. 2, p. 689-698.

Zoback, M. D., J. H. Healy, J. C. Roller, G. S. Gohn, and B. B. Higgins, 1978,Normal faulting and in situ stress in the South Carolina coastal plainnear Charleston: Geology, v. 6. p. 147-152.

39

TABLE 1. VERTICAL STRESS PROFILE FOR HOLE Ue-25pl

Depthinterval(m)

Averagedensity(g/cm3)

Source Sv albase(bars)

S abase)(bars)

Lithologic unit

00 - 39 1.8 1 6.88 6.25 Alluvium39 - 55 1.9 1 9.86 9.00 Ammonia Tanks member of

Timber Mountain Tuff55 - 79 2.1 1 14.80 13./5 [iva Canyon member of

Paintbrush Tuff79 - 101 2.12 2 19.37 18.86

101 - 122 2.11 3 23.71 23.40122 - 152 1.93 3 29.38 29.60 Topopah Springs member152 - 183 2.27 3 36.28 36.70 of the Paintbrush Tuff183 - 213 1.94 3 41.98 43.35213 - 244 2.00 3 48.06 49.77244 - 274 2.06 3 54.12 56.22274 - 305 2.21 3 60.83 63.19305 - 335 2.31 3 67.62 70.29335 - 367 2.21 3 74.33 77.59367 - 396 1.92 3 79.79 83.41 Tuffaceous Beds of396 - 430 1.98 3 86.39 90.12 Calico Hills430 - 439 2.25 3 88.37 92.06 Bedded tuff439 - 451 1.98 3 91.86 95.93 Prow Pass member of457 - 482 2.16 3 97.15 101.55 Crater Flat Tuff482 - 488 2.15 2 98.41 102.95488 - 518 2.14 4 104.70 109.63518 - 546 2.05 4 110.33 115.49546 - 551 2.04 4 111.33 116.52 Bedded tuff551 - 519 2.16 4 116.98 122.63 Bullfrog member of579 - 610 2.34 4 124.09 129.98 Crater Flat Tuff610 - 640 2.34 4 130.97 136.96640 - 671 2.20 4 137.65 143.82671 - 684 2.11 4 140.34 146.62684 - 688 2.17 4 141.19 147.48 Bedded tuff688 - 701 2.13 4 143.90 150.2/ Tram member of Crafer701 - 732 2.17 4 150.49 157.18 Flat Tuff732 - 762 2.24 4 157.08 164.01762 - 792 2.22 4 163.61 170.85792 - 823 2.27 4 170.51 177.92823 - 853 2.31 4 177.30 184.74853 - 872 2.28 4 181.55 189.06

40

TABLE 1 (continued). VERTICAL STRESS PROFILE FOR HOLE Ue-25p1

Depthinterval(m)

Averagedensity(g/cm3)

Source Sv atbasel(bars)

Sv atbase2

(bars)

Lithologic unit

872 - 884 2.20 4 184.14 191.83 Lithic Ridge Tuff884 - 914 2.20 4 190.61 198.51914 - 945 2.21 4 197.32 205.48945 - 975 2.26 4 203.96 212.30975 - 1006 2.32 4 211.01 219.491006 - 1036 2.36 4 217.95 226.461036 - 1064 2.25 4 224.12 232.771064 - 1067 2.29 4 224.79 233.44 Bedded tufflub/ - 109/ 2.34 4 231.67 240.53 Older tuff1097 - 1128 2.38 4 238.90 247.881128 - 1137 2.46 4 241.07 250.041137 - 1170 2.55 4 249.32 258.49 Fanglomerate11/U - 119b 2.46 3 255.35 264.47 Older tuff1195 - 1231 2.46 2 264.03 273.751U31 - 1U98 2.74 2 282.02 291.36 Lone Mountain Dolomite1298 - 1311 2.71 5 285.47 294.891311 - 1341 2.76 5 293.58 302.951341 - 1372 2.74 5 301.90 311.281372 - 1402 2.72 5 309.90 319.351402 - 1433 2.74 5 318.22 327.681433 - 1463 2.76 5 326.33 335.751463 - 1494 2.75 5 334.68 344.081494 - 1524 2.75 5 342.77 350.741524 - 1554 2.71 5 350.74 360.481554 - 1585 2.75 5 359.09 368.971585 - 1615 2.77 5 367.23 377.181615 - 1634 2.77 5 372.39 382.391634 - 1646 2.79 5 375.67 385.68 Roberts Mountain1646 - 1676 2.74 5 383.73 393.89 Formation1676 - 1707 2.78 5 392.18 402.381707 - 1737 2.80 5 400.41 410.601737 - 1768 2.80 5 408.92 419.091768 - 1801 2.77 5 417.88 --

Density sources: (1)(2)(3)(4)(5)

Snyder and Carr, 1982Averages of values within this same unit, this holeBirdwell compensated density log, run Nov. 30, 1982Dresser-Atlas compensated density log, run Jan. 18, 1983Birdwell compensated density log, run May 3, 1983

Sv at base: 1 - derived from density sources listed above2 - derived from borehole gravity measurements of Healey et al.

(1984)

41

TABLE 2a. VERTICAL STRESS PROFILE FOR THE UPPER PART OF USW G-3

Log depthinterval(m)

True depthinterval(m)

Averagedensity(g/cm3)

Source Sv at base ofinterval (bars)A B

Lithologic unit

o - 129 same 2.1 1 26.55 28.60 Tiva Canyon member ofPaintbrush Tuff

129 - 431 same 2.12 2 89.09 91.41 Topopah Springs memberof the Paintbrush Tuff

431 - 459 same 1.95 2 94.63 95.63 Ash flow tuff459 - 476 same 2.12 2 98.16 98.54 Bedded tuff4/6 - 601 same 2.10 2 125.12 122.19 Prow Pass member of the

Crater Flat Tuffbo/ - 611 same 2.10 2 125.94 122.94 Bedded tuff611 - 716 same 2.23 2 162.00 -- Bullfrog member of the776 - 777 same 2.23 2 162.22 -- Crater Flat Tuff777 - 796 same 2.23 2 166.37 --

796 - 798 same 2.05 3 166.77 --

798 - 803 same 2.10 3 167.70 -- Bedded tuff803 - 823 803-822 1.98 3 171.47 -- Tram member of the823 - 853 822-852 2.09 3 177.61 -- Crater Flat Tuff853 - 884 852-882 2.18 3 184.02 --884 - 914 882-912 2.32 3 190.84 --

914 - 945 912-942 2.32 3 197.66 --945 - 975 942-972 2.29 3 204.39 --

975 - 1006 972-1002 2.30 3 211.15 --

1006 - 1036 1002-1031 2.17 3 217.32 --

1036 - 1067 1031-1061 2.18 3 223.73 --

1067 - 1097 1061-1090 2.22 3 230.04 --

1097 - 1128 1090-1120 2.24 3 236.63 --

1128 - 1158 1120-1149 2.30 3 243.17 --1158 - 1173 1149-1164 2.29 3 246.54 --

1173 - 1177 1164-1168 2.27 3 247.43 -- Bedded tuff

Sources of density information: (1) Snyder and Carr, 1982(2) Averages of values from

and USW H-1holes USW G-2, Ue-25pl,

(3) Dresser-Atlas compensated density log of USW G-3,run March 23, 1982

True depth intervals calculated from Eastman Whipstock gyro multishot survey of USW G-3,run on March 19, 1982. Combined run data used.

Sv at base of interval: (A) calculated from densities given in column 3(B) calculated from borehole gravity results of Healey

et al. (1984)

42

Table 2b. VERTICAL STRESS PROFILE FOR THE LOWER PART OF USW G-3

Log depth True depth Average Sv at Elevation Sv Correctedinterval Interval density* base of correction correction Sv

Cm) (m) (g/cm3) interval (m) (bars) (bars)(bars)

1177 - 1219 1168-1208 2.22 256.13 -4.88 -1.00 255.131219 - 1250 1208-1238 2.18 262.54 -12.50 -2.57 259.971250 - 1280 1238-1267 2.22 268.85 -17.07 -3.51 265.341280 - 1311 1267-1296 2.19 275.07 -23.16 -4.77 270.301311 - 1341 1296-1324 2.26 281.27 -26.21 -5.39 275.881341 - 1372 1324-1353 2.28 287.75 -29.26 -6.02 281.731372 - 1402 1353-1381 2.26 293.95 -35.36 -7.28 286.671402 - 1433 1381-1409 2.29 300.23 -41.45 -8.53 291.701433 - 1463 1409-1437 2.29 306.51 -47.55 -9.79 296.721463 - 1494 1437-1465 2.24 312.66 -53.64 -11.04 301.621494 - 1536 1465-1507 2.25 321.92 -68.88 -14.18 307.74

*Values from Dresser-Atlas compensated density log of USW G-3, run March 23, 1982.

Sv at base of interval calculated from densities in column 3. No boreholegravity measurements were made at this depth.

Elevation correction: difference in height of surface directly above base ofinterval, and elevation of top of hole. This is important in this part of USW G-3due to excessive hole drift. Hole drift used for true vertical depths arecalculated from the Eastman Whipstock gyro multishot survey of USW G-3, run onMarch 19, 1982. Combined run data used. Elevation corrections are calculatedfrom plotting horizontal projection of hole drift on a topographic base map. Mapcontour intervals are 20 feet; thus, we estimate + 3 m uncertainty in theelevation corrections listed.

43

TABLE 3. SUMMARY OF STRESS MEASUREMENTS, USW G-3 AND UE-25pl

Shut-in Hydro-Logged Breakdown pumping static Pore2

Hole depth pressurel pressure pressure pressure Sh Sv 3 T4 SH Comments(m) (bars) (bars) (bars) (bars) (bars) (bars) (bars) (bars)

G-3 1,074 79 68 105 31 68+2 206 13 107+10

G-3 1,338 173 115 131 56 115+2 258 59 175+9 Interpretation based onfirst two cycles only.

G-3 1,356 168 114 133 57 114+2 263 64 181+8

P1 1,564 none1 337 153 115 337+2 353 Reopening preexistingfracture

P1 1,573 236 207 154 116 207+2 356 41 310+11

Pal 1,693 none1 366 166 128 365+10 388 Reopening preexistingfracture

1. No clear breakdown pressures seen; see comments column and Figure 10.2. Based on water table at the appropriate depth: 385 m in Ue-25pl and 752 m in USW G-3.3. Based on densities and calculations in Tables 1 and 2, plus topographic averaging in USW G-3 (see text).4. T derived from observation of fracture reopening pressures on later cycles.

TABLE 4. LOWER LIMITS ON Sh FROM TESTS IN THE UPPER PART OF Ue-25pl (VOLCANIC UNITS)

Depth (m) Lithology Pfailure (bars) PP (bars) T (bars)* Lower limit on Sh (bars)

731 Tram member: ash flow, 189 34 54 85zeolitized, partly tomoderately welded

1,001 Lithic Ridge Tuff: ash 236 61 54 122flow, zeolitized, partlyto moderately welded

1,131 Older tuffs: argillized, 204 73 54 112calcitized, partly welded

c * Value of T based on observations listed in Table 5.

TABLE 5. CONSTRAINTS ON HYDRAULIC FRACTURING TENSILE STRENGTH (Thydrofrac) IN THE VOLCANIC UNITSAT YUCCA MOUNTAIN

Pb - Pb (T=O)Drill hole Depth () Lithologic unit Thydrofrac (bars) Borehole diameter (cm)

USW G-1 1,288 Lithic Ridge Tuff, partly 50 10.8welded

USW G-2 1,026 Bullfrog member; moderately 47 18welded

1,209 Bedded tuff below Tram member, 62 19vitric (partly argillic)

USW G-3 1,074 Tram member: ash flow tuff, 13 10.8devitrified, moderately welded

1,338 Lithic Ridge: ash flow tuff, 59 9devitrified, moderately welded

1,356 Lithic Ridge: ash flow tuff, 64 9devitrified, moderately welded

Ue-25pl 731 Tram member: ash flow tuff, 40 27.3welded, zeol itized

Thydrofrac is derived from the difference between the breakdown pressure, Pb, and the fracture

reopening pressure on the second cycle, Pb (T=O). Tests listed are those for which a clear

breakdown was observed. Tests believed to have reopened preexisting fractures are omitted.

TABLE 6. INFLUENCE OF BOREHOLE DEVIATIONS ON RESULTS OF HYDROFRAC TESTS

DepthHole (m)

Borehole orientation1

drift azimuth

Approximate magnitudes ofprincipal stresses (bars)2

Sv SH Sh Pp

Predicted3

hydrofracorientation

Predicted3

breakdownpressure(bars)

Ideal4

breakdownpressure(bars)

USW G-3 1,074

1,338

1,356

UE-25pl 731

1,001

1,131

1,564

1,573

1,693

14.5e

0

19"

20.5-

0

1

1.25"

3.250

3.250

70

N85.5*W

S83"W

S82.5 W

S25. SW

S2 E

S30OW

S32.5'E

S27E

S3.50E

206 -107 68 31

258 -175 115 56

263 -181 114 57

151 -127 73 34

211 -180 111 61

243 -208 129 73

353 -304 194 115

356 -310 207 116

388 -334 214 128

N25*E

N25eE

N25.5 E

N25eE

N25eE

N25eE

N25eE

N25aE

N25eE

153

144 161

87.5 97

92 92

153 153

170

179 179

277 278

310 311

306 308

1. Obtained from Eastman Whipstock deviation log of hole (see figure 10).2. Values of Sv and Pp. are calculated as described in text. Values of Sh are estimates obtained from hydraulic

fracturing tests. SH magnitudes are those calculated using standard hydrofrac equation (1) in text, or approxi-mate values obtained by interpolation of average stress ratios in these two drill holes plus USW G-1 and USW G-2.These values are used for obtaining an estimate of the effect of the hole deviation and do not represent actualSH values obtained from the tests, with the exception of the three USW G-3 values and the value at 1,573 m depthin Ue-25pl.

3. Values predicted assuming a far field stress orientation of Sv vertical, SH at N25 E, and Sh at N65 W and ahole deviation as given by the values in column 2. Predicted breakdown pressure is Pb (T=O), the value of theleast compressive stress in the borehole wall.

4. Value of Pb (T=O) obtained for a perfectly vertical hole under the same stress conditions, using equation (2)of the text.

A P P E N D I X I

USW G-3 Televiewer Photos

TELEVIEWER LOG DATA SHEET - USW G-3 (Part I)

Date logged: November 18 and 19, 1983Tool no.: 2Surface panel no.: 1Zero depth at: center of boot at top of casingTop of logged interval: 2,600' (bottom of casing)Bottom of logged interval: 2,761' (tool would not

Rezero: -5.2'

fall below this depth)

Depth interval(ft)

2,6002,6152,6602,6752,8002,8102,8803,0153,0303,2303,2803,5103,5653,5903,6053,6253,6753,6953,720

- 2,615- 2,660- 2,675- 2,800- 2,810- 2,880- 3,015- 3,030- 3,230- 3,280- 3,510- 3,565- 3,590- 3,605- 3,625- 3,675- 3,695- 3,720- 3,761

Boreholegai n

5454434345544445555

Blanking TV Gain

044044044044044044044044044044047047047044045044044044038

230230230230257257214214214214234245230230244244221234264

Logged by: L. Mastin, T. Denham, J. Stock

48

2650I

k,

L.I I

I f.,

; 2700Iil

.;I

I

'-I

i

1

i-I

of 276

'I, 2t

200

NTS USW G- 3

49

2750 _ j2800 t2 1250

277522527

F II -

280 U 8 5 0 290

NTS USW G- 3

50

, , ,- � �-, I I I 11 - . . . - .. I � I �rk I :1 -,

I200 _1290 13000

250 1-3000

3050

NTS USW G- 3

51

.: ! .~~~1I

__*303

._. .__ . ._^ . a_ . . _ _ _ .

~~~~I .G

.'h.0'''.

{ 0171

I

3075

- 3100

;3,.26It2

, -:

I';A

. .-.

l 3 150,: . Iio.320

NTS USW G 3

52

TELEVIEWER LOG DATA SHEET - USW G-3 (Part II)

Date logged: December 7, 1983Tool no.: 2Surface panel no.: 1Zero depth at: center of boot at top of casingTop of logged interval: 3,280'Bottom of logged interval: 4,480' (hole blocked)

Rezero: -6.4'

Depth nterval(ft)

3,2803,3103,3353,3413,7603,8803,8954,2004,410

- 3,310- 3,335- 3,341- 3,760- 3,880- 3,895- 4,200- 4,410- 4,480

Boreholegai n

444343456

047047047047047047047047047

Blanking TV Gain

238280400400400400400400400

Logged by: J. Stock, L. Mastin, J. Springer

53

i

3200 t3250 3300

I I E1

I &_teI

'3300 ~ ~ ~ ~ ~ ~ 38

NTS USW G- 3

54

it

3360 a3400 ' 3450

13375 3425 ,'3476

.30 3500a

.~~~~~~~~T USW G- 3

55

l -l i3500 3000

3625 - 1 35? 5 | 3 0 2~~~~~~~~~~~385 ,

3650 - . : 3eooew S ~~~~~~3 3e0

NTS USW G- 356

3650 3700 3750

| I _ ! ; 1d I

3675 -3725 3

3700 3750 3800

NTS USWG-3

57

38003850 3900

.

.Q

II

111t

3825 3ars 5 3925

I I

38ss '3. S900 r A 3950

NTS USW G. 3

58

3950

4 8976

III L-

J

i

i

4000

.I.!III

I J 4026

.

i

406

�I

I

I

iI

i1.

l

4060

I

I

.1

i,

- 0i

.

i

l

i

410

4000

NTS USW G 3

59

I

I -4100 4150 14200

NTS USW I 3

60

4250 4300 4350

4276 4325 4376

4300 4350

NTS USW G-

61

4400 4450

4425 4475

-' 4480

4450

NTI USW G- 3

62

A P P E N D I X I I

Ue-25pl Televiewer Photos

TELEVIEWER LOG DATA SHEET - Ue-25pl (Part I)

Date logged: February 28, 1983Tool no.: 1Surface panel no.: 1Zero depth at: center of boot at rig floorTop of logged interval: 1,625' (bottom of casing)Bottom of logged interval: 3,945' (hole blocked)

Depth interval Borehole Blanking TV Gain(ft) gain

1,575 - 1,600 8 0591,600 - 1,675 8 0621,675 - 1,695 10 0621,695 - 1,715 7 0621,715 - 1,860 8 0621,860 - 1,905 8 0591,905 - 1,955 7 0591,955 - 2,000 7 0612,000 - 2,030 7 0492,030 - 2,425 6 0492,425 - 2,625 8 0492,625 - 2,750 7 0492,750 - 2,920 8 0492,920 - 2,985 8 0452,985 - 3,390 8 0433,390 - 3,440 8 0423,440 - 3,455 8 0453,455 - 3,650 8 0443,650 - 3,915 8 0513,915 - 3,945 8 Oxx

Logged by: K. Brown, J. Healy, J. Stock, G. Zwart

63

157515 S Is

lots1600

1825

'-Ills

1725Iis

64

ls1726 1825

1776

less

tIIH

Il$$

IV S

65

loll'III

1878

1ma

I I 2

lIs5

311

ta

'Si'

66

35112025

less ties

I2

list

2e asI

67

2175222 t atI

22is 2213

2 21 1

2313

222S 2325

68

2325 2 3 1 1

i

T

I

i

I

i

i

iL

i

i

II

Ii

.i

2 351 2313 ~~~~~~~~2463

242 5

243

2)ws 2475

69

247525tM

2S35 III$251

25112S2 5

2525

70

2625 is s 272*

Till l1ilist

2712I'll

275

71

75

Ue25 p-172

2925 2115lI M

tug7

so3Is

site

2975 Acsi

73

3075 3 12 5

s188 a1lu

317 322 '

74

I2 532153225

1211 life 381

3 11X3253 2 1 5

75

3 416

3421 41

33S5

41

3421

3nilsits

3411$ft S

76

3525 sills 3 ItI

I

-. 2133215t

3 1 I 5

ISIS

251

61

77

X7"X 171%3 675

3113f

3115

3800

sit$3725

78

I

I

I

Ue25p-1

79

TELEVIEWER LOG DATA SHEET - Hole Ue-25pl (Part II)

Date logged: July 17, 1983Tool no.: 1Surface panel no.: 1Zero depth at: top of casing, 1' above site grade Rezero: -3.1'Top of logged interval: 4,260' (bottom of liner in hole)Bottom of logged interval: 5,000' (nearing end of wireline)

Depth interval(ft)

4,255 - 4,2754,275 - 4,3404,340 - 5,000

Boreholegai n

768

Blanking TV Gain

038038038

232232232

Logged by: . Svitek, J. Stock

TELEVIEWER LOG DATA SHEET - Hole Ue-25pl (Part III)

Date logged: November 16, 1983Tool no.: 2Surface panel no.: 1Zero depth at: center of boot to top of casing Rezero:Top of logged interval: 4,975' (match up and overlap 25' withBottom of logged nterval: 5,900' (holefilled to this depth)

-3.9'previous log)

Depth interval(ft)

4,975 - 5,0755,075 - 5,8605,860 - 5,900

Boreholegain

745

Blanking TV Gain

047047047

232232248

Logged by: J. Stock, L. Mastin, T. Denham

80

4300,

4255i

4350

1

4375, 43254275

4300 4350 4400

4450

44004500

4425

4450

4475

4500

4525

4550

82

4550 4600

4625

4650

4650

4575

4600

4675

4700

83

4750 48004700

4725

A7l 4850 I

I~

e

.bm

A-. SDw~

84

48504900

4925

4950

4950

4975

5000

4875

4900It

85

50005050 5100

5025 | D~o5075 5125

5050 I @' S'sw

100 ' '_^L5 150 £* t

Ue25p. 1

86

5150 5200 5250

5175 5

5200 5250 5300

Ue25p-

87

5300 5350 5400

5325 -3 5323 | 1 53t5537 5425

5350 a 1 5400 5450

Ue25P- 1

88

5450 55006550,

k

6476 5525

6I60

655

6575

II

I

5500

.

5600

Ue25p- 1

89

5600 5650 5700

5625 | -1 5875 5725

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I'

5650 5700 5750

Ue25p. 1

90

50 350

575;775

5800Ue25 p-1

914 U.S. GOVERNMENT PRINTING QfFICE:1987- -I I7